Long-term mobile robot localization in dynamic environments using spectral maps

Krajnik, Tomas and Pulido Fentanes, Jaime and Martinez Mozos, Oscar and Ekekrantz, Johan and Hanheide, Marc and Duckett, Tom (2015) Long-term mobile robot localization in dynamic environments using spectral maps. [Video]

Full content URL: https://www.youtube.com/watch?v=Qw1kS_5zVwE

Audio/Video Files
Click on one of the thumbnails to start playing a video. Note that you need to login in order to preview restricted items.
AAAI-fremen-final.mp4
[Download]
1/1AAAI-fremen-final.mp4
[img] Video (MP4)
AAAI-fremen-final.mp4 - Whole Document

77MB
Item Type:Video
Item Status:Live Archive

Abstract

The video presents a novel approach for vision-based topological localisation in dynamic indoor environments. In contrast to other approaches that rely on static image features, our method explicitly models the temporal dynamics of the visibility of image features. The proposed spatio-temporal world model is able to predict the visibility of image features at different times of day, which allows to construct time-specific environment appearance models. This approach improves the robot localisation capabilities during long-term operations.

Keywords:long-term autonomy, visual localisation, mobile robotics
Subjects:H Engineering > H670 Robotics and Cybernetics
Divisions:College of Science > School of Computer Science
ID Code:17950
Deposited On:23 Jul 2015 10:32

Repository Staff Only: item control page